Eur J Cardiothorac Surg 2001;20:58-64
© 2001 Elsevier Science NL
Rewarming from accidental hypothermia by extracorporeal circulation.
A retrospective study
M. Farstada,
K.S. Andersenb,
M.-E. Kollera,
K. Grongc,
L. Segadalb,
P. Husbya
a Department of Anesthesia and Intensive Care, Haukeland University Hospital, University of Bergen, N-5021 Bergen, Norway
b Department of Heart Disease, Haukeland University Hospital, University of Bergen, N-5021 Bergen, Norway
c Surgical Research Laboratory, Haukeland University Hospital, University of Bergen, N-5021 Bergen, Norway
Received 12 January 2001;
received in revised form 23 March 2001;
accepted 23 March 2001.
Corresponding author. Tel.: +47-5597-6865; fax: +47-5597-6898
e-mail: phus{at}haukeland.no
 |
Abstract
|
|---|
Objective: Twenty-six patients with accidental hypothermia combined with circulatory arrest or severe circulatory failure were rewarmed to normothermia by use of extracorporeal circulation (ECC). The aim of the present study was to evaluate our results. Patients and methods: The treatment of six female and 20 male patients (median age: 26.7 years; range 1.976.3 years) rewarmed in the period 19872000 was evaluated retrospectively. Hypothermia was related to immersion/submersion in cold water (n=17), avalanche (n=1) or prolonged exposure to cold surroundings (n=8). Prior to admission, the trachea was intubated and cardiopulmonary resuscitation (CPR) initiated in all patients with cardiorespiratory arrest (n=22), whereas in those with respiration/circulation (n=4) only oxygen therapy via a facemask was given. Results: Nineteen of the 26 patients were weaned off ECC whereas seven died because of refractory respiratory and/or cardiac failure. Eight of the 19 successfully weaned patients were discharged from hospital after a median of 10 days. One patient died 3 days after circulatory arrest (complete atrioventricular block) resulting in severe cerebral injury. The remaining ten patients died following 12 days due to severe hypoxic brain injury (n=5), cerebral bleeding (n=1) or irreversible cardiopulmonary insufficiency (n=4). Based on the reports from the site of accident, two groups of patients were identified: the asphyxia group (n=15) (submersions (n=14); avalanche accident (n=1)) and the non-asphyxia group (n=11) (patients immersed or exposed to cold environment). Seven intact survivors discharged from hospital belonged to the non-asphyxia group whereas one with a severe neurological deficit was identified within the asphyxia group. Conclusion: Patients with non-asphyxiated deep accidental hypothermia have a reasonable prognosis and should be rewarmed before further therapeutic decisions are made. In contrast, drowned patients with secondary hypothermia have a very poor prognosis. The treatment protocol under such conditions should be the subject for further discussion.
Key Words: Accidental hypothermia Near-drowning Drowning Extracorporeal circulation Extracorporeal blood rewarming
 |
1. Introduction
|
|---|
Severe accidental hypothermia remains a challenging clinical problem mainly affecting victims of outdoor accidents who suffer prolonged exposure to cold environment or immersion/submersion in cold water. Most reports from an urban environment are usually associated with drug or alcohol abuse, serious illness in elderly or debilitated patients, suicide, occupational disaster or accidents related to sport activity [1]. Deep accidental hypothermia (core temperature below 28°C) causes circulatory and neurological disturbances; even with immediate treatment, high mortality rates are reported [1]. The hypothermia literature includes many anecdotal reports and most often single or only few cases are analyzed. Different patient characteristics, associated conditions and aetiology have resulted in variable interpretations and recommendations. Still the number of reports presenting larger patient-populations of accidental deep hypothermia is small [27]. Several reports have advocated cardiopulmonary bypass (CPB) as the method of choice for rewarming from severe accidental hypothermia associated with a persistent non-perfusing rhythm (asystole or ventricular fibrillation) [8].
To our knowledge, a general consensus is still not available on the optimal treatment protocol in cases of drowned/near-drowned subjects suffering secondary hypothermia. The use of CPB resuscitation in such patients is discussed on a limited scale in the literature, but should be subjected to more extensive and critical evaluation.
After 1987, 26 patients admitted with hypothermia combined with circulatory arrest or cardiorespiratory insufficiency were rewarmed by extracorporeal circulation (ECC) as previously described [9]. We here present a retrospective analysis of 26 patients treated by ECC in the period 19872000.
 |
2. Patients and methods
|
|---|
Twenty-six hypothermic patients with cardiac arrest or severe circulatory failure were treated during 19872000. Soon after admission, arterial and central venous lines were established during ongoing cardiopulmonary resuscitation (CPR) and transfer to the operating theatre. Prior to start of and early during rewarming pH was adjusted to 7.157.20. Acidbase parameters were not corrected to actual temperature. Rapid rewarming on extracorporeal circulation was initiated immediately after admission using standard equipment for open-heart surgery (heparin-uncoated oxygenator/tubes). General anesthesia was performed with fentanyl, midazolam and pancuronium according to body weight and supplemented with isoflurane in oxygen/air via a ventilator (Servo 900 C; Elema-Siemens, Sweden).
Arterial and venous access for ECC was performed via the femoral artery and vein in 21 patients or via sternotomy in the ascending aorta and in the right atrium in five patients, in two of them due to inadequate venous return and in the remaining three due to low age and the surgeons preference.
All patient data presented in this report were collected retrospectively from transport records, hospital journals, laboratory records and surgeons reports. For some of the presented variables, the data are incomplete (laboratory records: analysis not requested).
2.1. Statistics
Data were analyzed with Minitab® v. 10.1. Continuous variables were investigated for normal distribution by KolmogorovSmirnov tests for normal distribution and Barlett tests for homogenous variance between groups. If these conditions were not met (P<0.05) variables are presented as median with first and third quartile (Q1, Q3) in parenthesis. Data with normal distribution are given as mean±standard deviation (SD). Group differences were studied by two-sample t-tests or MannWhitney tests for data with or without normal distribution, respectively. Group differences for nominal scale data are studied by Chi-square tests. Data available for all patients were included in multiple regression analyses (step-down procedure). For all analyses, the level of significance was set to P<0.05.
 |
3. Results
|
|---|
At time of rescue, insufficient respiration and circulation were diagnosed either by paramedics or air ambulance doctors at the scene of accident, in 18 of the 26 patients (Table 1). In these the trachea was intubated and advanced CPR initiated and continued during transport (air ambulance) to our emergency department (ED). In the remaining eight patients vital signs were present at time of rescue. During early handling and transportation four of them developed cardiorespiratory arrest and were treated as described. In the four remaining patients with intact ventilation and circulation oxygen was delivered via a facemask and their clinical condition evaluated continuously. After arrival in the ED one patient with intact circulation converted to ventricular fibrillation whereas another developed circulatory arrest early during external rewarming in the intensive care unit (ICU). Cardiopulmonary resuscitation time (CPR-time) until start of ECC was 150±88 min ranging from 0 to 295 min.
Admission characteristics of all hypothermic patients treated on extracorporeal circulation are given in Tables 1 and 2. The 26 patients (six female/20 male), median age 26.7 years (3.8, 44.5) (range 1.976.3 years), with rectal temperature of 23.3±3.3°C (range 17.331°C) were either found immersed/submersed (n=17), outdoor in cold environment (n=8) or in an avalanche (n=1) (Table 1). Acid base parameters and s-potassium values upon admission are presented in Table 2. In only a few patients corrective treatment of acid-base disturbances were initiated prior to admission. Ethanol intoxication was found in eight of the 26 patients. Electrocardiographic examination at time of admittance revealed either cardiac arrest (n=21) or bradyarhythmia (n=5). Extracorporeal rewarming (gradient 5°C, heat exchanger vs. actual core temperature) was completed after 106±34 (40.0186.0) min. Mean arterial pressure during ECC (n=23) was 50 mmHg ranging from 2080 mmHg.
Analysis of every single patients history indicated that 57% of the victims (n=15) most probably underwent asphyxia before or during cooling. All of these patients except one (avalanche accident) were submersions with cardiorespiratory arrest at time of rescue. In the remaining patients (n=11) related to situation at time of rescue asphyxia was unlikely (Table 1). Based on this information two subgroups of patients were identified and further analyzed: the asphyxia group (n=15) and the non-asphyxia group (n=11) (Table 3). Information on duration of asphyxia and hypothermia were available in few patients and are not presented. Transportation time to hospital were also partly unknown, but for all patients far above 30 min.
View this table:
[in this window]
[in a new window]
|
Table 3. Data at hospitalization for the 15 patients with presumed primary asphyxia and the 11 patients with presumed primary hypothermia according to the recordsa
|
|
Age-differences were seen between the two groups (P<0.001) where a major part of the asphyctic patients were paediatric submersion cases (Table 3). Furthermore the asphyxia group patients underwent significantly longer CPR (P<0.05), showed a higher frequency of asystole (P<0.005), had a more pronounced acidosis (P<0.01) and accordingly a higher s-potassium level (P<0.05).
The results after rewarming are presented in Table 4. Seven of the 26 patients could not be weaned off ECC due to irreversible cardiac arrest or severe cardiopulmonary insufficiency resistant to therapy. The treatment was here ended in the operating theatre. Post-rewarming pulmonary edema was seen in only one of the survivors. None of the preoperative variables like age, sex, eventual intoxication, asphyxia, temperature, cardiac rhythm, CPR-time, acid-base-parameters or serum-electrolytes could significantly predict if circulation could be re-established or not. In the remaining 19 patients spontaneous circulation could be restored. During the following 2448 h ten of the 19 patients transferred on ventilator to the ICU died (Table 4). Of the nine remainders, eight could finally be discharged home. Seven of them belonged to the non-asphyxia group and survived without any sequelae, whereas one identified within the asphyxia group survived, but suffered from a severe neurological deficit (Table 4).
Of the eight patients with intact cardiorespiratory function at time of rescue, six developed cardiorespiratory arrest prior to or after admission. Professional CPR was initiated immediately and five of the six were long-term survivors.
Significant differences were also present for asphyxia or not, sex, age, CPR-time, intoxication and initial cardiac rhythm between the non-survivors (n=18) and those discharged alive (n=8) as presented in Table 5. By use of multiple regression analysis preoperative factors proposed to influence the final outcome, were tested. Factors significantly predicting a negative outcome in this patient population were (a) assumed asphyxia (P=0.019) and (b) prolonged CPR-time (P=0.022).
 |
4. Discussion
|
|---|
Clinical experience have demonstrated that young, healthy people can survive accidental deep hypothermia with no or minimal cerebral impairment even after prolonged circulatory arrest [10]. Severe accidental hypothermia is still associated with a high mortality rate ranging from 3080%. Accordingly, Vretenar and co-workers [8] reported an overall mortality of 40% in their collective review of 68 severely hypothermic patients resuscitated with the use of ECC whereas Kornberger and Mair [4], apparently in a similar group of patients, observed an overall mortality rate of 85% corresponding to that reported following the Mt. Hood Tragedy [3]. Walpoth and co-workers [7] reported a 65% survival of rewarmed hypothermic victims with cardiac arrest. That is twice the overall survival rate presented in this study. Differences in accident mechanisms, patient populations and pre-hospital handling as well as indications for rewarming by ECC may explain the great variability of survival from one study to another.
When analyzing our two groups of patients, a 63% survival rate was present in the non-asphyxia group, identical to [7], but in sharp contrast to the poor outcome (5%) seen in the asphyxia group.
A major part of the patients in our asphyxia group are paediatric near-drowning/drowning cases. Although a few reports during the last decades have revealed a small, but significant number of paediatric near-drowning victims who recovered without major neurological sequelae after arriving in the ED in a comatose state [1113], most authors describe a dismal prognosis of near-drowning/drowning patients admitted with circulatory arrest [13,14]. The low median age in the non-survivor group compared to median age of survivors (Table 5) is explained by the fact that a major part of our patients were near-drowning/drowning cases, and not by the age per se as negative predictor. High age as negative predictor for survival has been demonstrated earlier [2]. In our limited sample of 26 patients the univariate analysis of age and several other grouping factors (sex, intoxication, CPR-time and cardiac rhythm) indicate a significant influence of these factors on outcome. However, in the present patient population many of these factors are not independent predictors of outcome but are negatively or positively intercorrelated. Taking this into account, the multiple regression analysis demonstrating that a history of asphyxia and prolonged cardiopulmonary resuscitation time as significant negative predictors for outcome is in accordance with previous findings describing a clear reduction of survival associated with asphyxia and with circulatory arrest requiring resuscitation [2]. Peterson [15] found no neurological intact survivors in subjects requiring CPR after admission to the emergency department. Most studies report a significant incidence of severely neurologically impaired survivors among patients requiring prolonged CPR [14]. All our patients present within the asphyxia group were admitted to our ED after prolonged CPR and still in cardiorespiratory arrest. The only survivor in this group was a 3.7-year-old girl, playing in cold water for about 60 min prior to a submersion of more than 1 h. She probably survived due to hypothermic protection at time of submersion. She was discharged home after a prolonged hospital stay, however with a major cerebral sequelae. According to the presented results the dilemma of when and whom to resuscitate and by which means remains unclear and a general consensus is still not available.
Hypothermia by itself increases the microvascular fluid extravasation [16,17]. Similarly, use of crystalloid hemodilution (acetated Ringers solution) during ECC rewarming may affect the microvascular fluid shift with increased extravasation of fluid from circulation to the interstitial space. A marked cerebral swelling has been demonstrated following hypothermic as well as normothermic ECC [18,19]. It can be speculated if the cerebral swelling due to and following 1: ECC and 2: asphyxia, when present simultaneously may be of additive importance affecting clinical outcome. In our cases the ECC circuit was often primed with crystalloids (acetated Ringers solution). Addition of colloids or colloid priming [20,21] combined with ultrafiltration [22] will reduce fluid accumulation during ECC rewarming. Such measures could thus be beneficial when rewarming pediatric near-drowning/drowning victims on ECC.
Asphyxia turned out to be a significant negative predictor for patient outcome. A substantial part of the patients within the asphyxia group represented paediatric submersion cases where asphyxia was proposed to precede hypothermia. The poor outcome in these patients was therefore mainly due to the irreversible ischemic brain damages in primary asphyxiated victims. This experience is also in accordance with Locher et al. [2]. They found in the group of hypothermic patients with circulatory arrest following asphyxia a mortality rate of 100%. In the non-asphyxia group there was initially eight survivors with seven long-term survivors. One survivor was discharged to general ward and suffered a complete atrioventricular block on day 5 postoperatively. Despite CPR/pacemaker treatment he died on day 8 from cerebral hypoxic injury following the latter circulatory arrest. Another victim from the same group died 2 days after rewarming due to a cerebral haemorrhage that initially was unrecognised.
Prognostic markers such as potassium level >10 mmol/l, pH <6.5 and low arterial oxygen tension have been suggested to predict outcome. Such predictors may be helpful in making decisions if treatment should be initiated or not. Four of our patients had a s-potassium level above 10 mmol/l and they were all non-survivors. Schaller et al. [23] concluded that extreme hyperkalemia during acute hypothermia appears to be a reliable marker of death. Accordingly, they considered withholding resuscitation when serum potassium levels exceeded 10 mmol/l. Other reports [3,24] describe similar findings. In our study s-potassium was significantly higher in the asphyxia group compared with the non-asphyxia group. Despite the poor outcome of the patients within the asphyxia group, extreme hyperkalemia was present in only four patients, however contributing to the group differences.
The lowest oxygen tension (PaO2=1.8 kPa at 24°C (uncorrected for temperature)) was present in one of the victims within the non-asphyxia group. This patient survived, and was discharged without any sequelae.
Arterial pH differed significantly between the asphyxia group and the non-asphyxia group. However, while the mean pH in the asphyxia group was 6.72, the lowest pH of a survivor in our study was 6.47. Recently, Gilbert et al. [25] reported survival without major sequelae of a young woman rewarmed from a core temperature of 13.7°C and a pH of 6.54. This demonstrates the difficulty of taking decisions to continue or terminate resuscitation based on laboratory parameters.
The use of health care resources and costs in the non-survivors is limited (Table 5) supporting the view that all patients should benefit from any doubt, i.e. being rewarmed before further decisions are taken unless there are clear evidence of severe and prolonged asphyxiation.
 |
5. Conclusion
|
|---|
- The value of laboratory parameters as an aid of assessing hypothermic victims is according to our data limited. Extreme hyperkalemia (s-potassium >10 mmol/l) as a sign of cellular damage indicates a dismal prognosis.
- Circulatory arrested hypothermic patients with no hypoxic event preceding cooling and patients with present vital signs at time of rescue have a reasonable prognosis.
- Near-drowning/drowning victims admitted to the ED in a comatose state with cardiorespiratory arrest requiring prolonged CPR have a dismal prognosis.
- Victims with an indisputable history of asphyxia prior to cooling should not be rewarmed by CPB.
 |
References
|
|---|
-
Lønning P.E., Skulberg A.,
byholm F. Accidental hypothermia. Review of the literature. Acta Anaesthesiol Scand 1986;30:601-613.[Medline]
-
Locher T., Walpoth B., Pfluger D., Althaus U. Akzidentelle Hypothermie in der Schweiz (19801987) - Kasuistik und prognostische Faktoren. Schweiz Med Wochensch 1991;121:1020-1028.
-
Hauty M.G., Esrig B.C., Hill J.G., Long W.B. Prognostic factors in severe accidental hypothermia: experience from the Mt. Hood tragedy. J Trauma 1987;27:107-112.
-
Kornberger E., Mair P. Important aspects in the treatment of severe accidental hypothermia: the Innsbruck experience. J Neurosurg Anesthesiol 1996;8:83-87.[Medline]
-
Danzl D.F., Pozos R.S., Auerbach P.S., Glazer S., Goetz W., Johnsen E., Jui J., Lilja P., Marx J.A., Miller J., Mills W., Nowak R., Shields R., Vicario S., Wayne M. Multicenter hypothermia survey. Ann Emerg Med 1987;16:1042-1055.[Medline]
-
Pillgram-Larsen J., Svennevig J.L., Abdelnoor M., Fjeld N.B., Semb G., Østerud A., Skulberg A. Aksidentell hypotermi - risikofaktorer hos 29 pasienter med kroppstemperatur
30°C. Tidsskr Nor Lægeforen 1991;111:180-183.
-
Walpoth B.H., Locher T., Leupi F., Schüpbach P., Müleman W., Althaus U. Accidental deep hypothermia with cardiopulmonary arrest: extracorporeal blood rewarming in eleven patients. Eur J Cardio-thorac Surg 1990;4:390-393.[Abstract]
-
Vretenar D.F., Urchel J.D., Parrot J.C.W., Unruh H.W. Cardiopulmonary bypass resuscitation for accidental hypothermia. Ann Thorac Surg 1994;58:95-98.
-
Husby P., Andersen K.S., Owen-Falkenberg A., Steien E., Solheim J. Accidental hypothermia with cardiac arrest: complete recovery after prolonged resuscitation and rewarming by extracorporeal circulation. Int Care Med 1990;16:69-72.[Medline]
-
Walpoth B.H., Walpoth-Aslan B.N., Mattle H.P., Radanov B.P., Schrot G., Schaeffler L., Fischer A.P., von Segesser L., Althaus U. Outcome of survivors of accidental deep hypothermia and circulatory arrest treated with extracorporeal blood warming. N Engl J Med 1997;337:1500-1505.[Abstract/Free Full Text]
-
Siebke H., Breivik H., Rød T., Lind B. Survival after 40 min submersion without cerebral sequelae. Lancet 1975;1:1275-1277.[Medline]
-
Bolte R.G., Bleck P.G., Bowers R.S., Thorne J.K., Corneli H.M. The use of extracorporeal rewarming in a child submersed for 66 min. J Am Med Assoc 1988;260:377-379.[Abstract/Free Full Text]
-
Krandik G., Mantel K. Ertrinkungsunfälle im Kindesalter. Fortschr Med 1990;28:527-530.
-
Nichter M.A., Everett P.B. Childhood near-drowning: is cardiopulmonary resuscitation always indicated?. Crit Care Med 1989:993-995.
-
Peterson B. Morbidity of childhood near-drowning. Pediatrics 1977;59:364-370.[Abstract/Free Full Text]
-
Chen R.Y., Chien S. Plasma volume, red cell volume and thoracic duct lymph flow in hypothermia. Am J Physiol 1977;233:H605-H612.
-
Heltne J.K., Koller M.E., Lund T., Bert J., Rynning S.E., Stangeland L., Husby P. Dynamic evaluation of fluid shifts during normothermic and hypothermic cardiopulmonary bypass in piglets. Acta Anaesthesiol Scand 2000;44:1220-1225.[Medline]
-
Harris D.N., Oatridge A., Dob D., Smith P.L.C., Taylor K.M., Bydder G.M. Cerebral swelling after normothermic cardiopulmonary bypass. Anesthesiology 1998;88:340-345.[Medline]
-
Harris D.N., Bailey S.M., Smith P.L., Taylor K.M., Oatridge A., Bydder G.M. Brain swelling in first hour after coronary artery bypass surgery. Lancet 1993;342:586-587.[Medline]
-
Haneda K., Sato S., Ishizawa E., Horiuchi T. The importance of colloid osmotic pressure during open heart surgery in infants. Tohoku J Exp Med 1985;147(1):65-71.[Medline]
-
Cooper J.D., Maeda M., Lowenstein E. Lung water accumulation with acute hemodilution in dogs. J Thorac Cardiovasc Surg 1975;69(6):957-965.[Abstract]
-
Pouard, P.H. Ultrafiltration in children undergoing cardiopulmonary bypass. Proceedings, 21st Annual meeting of SCA (Society of Cardiovascular Anesthesiologists, Chicago. 1999; 414436.
-
Schaller M.D., Fisher A.P., Perret C.H. Hyperkalemia, a prognostic factor during acute severe hypothermia. J Am Med Assoc 1990;264:1842-1845.[Abstract/Free Full Text]
-
Roeggla G., Roeggla M., Wagner A., Hoedl W. Prognostic markers in patients with severe accidental hypothermia (letter). Resuscitation 1994;28:72-73.[Medline]
-
Gilbert M., Busund R., Skagseth A., Nilsen P.
., Solbø J.P. Resuscitation from accidental hypothermia of 13.7°C with circulatory arrest. Lancet 2000;355:375-376.[Medline]
This article has been cited by other articles:

|
 |

|
 |
 
W G Headdon, P M Wilson, and H. R Dalton
The management of accidental hypothermia
BMJ,
June 10, 2009;
338(jun10_2):
b2085 - b2085.
[Full Text]
|
 |
|

|
 |

|
 |
 
A. Alfonzo, A. Lomas, I. Drummond, and E. McGugan
Survival after 5-h resuscitation attempt for hypothermic cardiac arrest using CVVH for extracorporeal rewarming
Nephrol. Dial. Transplant.,
March 1, 2009;
24(3):
1054 - 1056.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
M. P. Anstadt and J. E. Lowe
Cardiopulmonary Resuscitation
Card. Surg. Adult,
January 1, 2008;
3(2008):
487 - 506.
[Full Text]
|
 |
|

|
 |

|
 |
 
E. Ruttmann, A. Weissenbacher, H. Ulmer, L. Muller, D. Hofer, J. Kilo, W. Rabl, B. Schwarz, G. Laufer, H. Antretter, et al.
Prolonged extracorporeal membrane oxygenation-assisted support provides improved survival in hypothermic patients with cardiocirculatory arrest
J. Thorac. Cardiovasc. Surg.,
September 1, 2007;
134(3):
594 - 600.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
Part 10.4: Hypothermia
Circulation,
December 13, 2005;
112(24_suppl):
IV-136 - IV-138.
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
Part 4: Advanced Life Support
Circulation,
November 29, 2005;
112(22_suppl):
III-25 - III-54.
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
M. P. Anstadt and J. E. Lowe
Cardiopulmonary Resuscitation
Card. Surg. Adult,
January 1, 2003;
2(2003):
471 - 494.
[Full Text]
|
 |
|